Comparison of ultraviolet absorbance and NO-chemiluminescence for ozone measurement in wildfire plumes at the Mount Bachelor Observatory

Abstract The goal of this paper is to evaluate the accuracy of the commonly used ozone (O 3 ) instrument (the ultraviolet (UV) photometer) against a Federal Reference Method (Nitric Oxide –chemiluminescence) for ozone measurement in wildfire smoke plumes. We carried out simultaneous ozone measurement with two UV O 3 photometers and one nitric oxide–chemiluminescence (NO-CL) ozone detectors during wildfire season (Aug. 1–Sept. 30) in 2015 at the Mount Bachelor Observatory (MBO, 2763 m above mean sea level, Oregon, USA). The UV O 3 shows good agreement and excellent correlation to NO-CL O 3 , with linear regression slopes close to unity and R 2 of 0.92 for 1-h average data and R 2 of 0.93 for O 3 daily maximum 8-h average (MDA8). During this two-month period we identified 35 wildfire events. Ozone enhancements in those wildfire plumes measured by NO-CL O 3 and UV O 3 monitors also show good agreement and excellent linear correlation, with a slope and R 2 of 1.03 and 0.86 for O 3 enhancements (ΔO 3 ) and 1.00 and 0.98 for carbon monoxide (CO)-normalized ozone enhancement ratios (ΔO 3 /ΔCO), respectively. Overall, the UV O 3 was found to have a positive bias of 4.7 ± 2.8 ppbv compared to the NO-CL O 3 . The O 3 bias between NO-CL O 3 and UV O 3 is independent of wildfire plume tracers such as CO, particulate matter (PM 1 ), aerosol scattering, and ultrafine particles. The results demonstrate that the UV O 3 absorbance method is reliable, even in highly concentrated wildfire plumes.